Nvidia Corporation

Pekka Janis

Minwoo Park

Tommi Koivisto

David Nister

Sangmin Oh

Zhaoting Ye

Yilin Yang

Daniel Herrera Castro

Bala Siva Sashank Jujjavarapu

John R Zedlewski

Zuoguan Wang

Gordon Grigor

Miguel Sainz

Timo Roman

Davide Onofrio

Chia-Chih Chen

Carolina Parada

Vijay Chintalapudi

Tero Kuosmanen

Yifang Xu

Xin Liu

Mehdi Sajjadi Mohammadabadi

Ozan Tonkal

Jan Nikolaus Fritsch

I-Kuei Chen

Yue Wu

Cheng-Chieh Yang

Nizar Assaf

Sriya Sarathy

William Zhang

Colin Tracey

Kari Hamalainen

Xin Tong

Jarmo Lunden

Tayfun Gurel

Bala Siva Jujjavarapu

Joonas Haapala

Apparatuses, systems, and techniques to identify objects with in an image. In at least one embodiment, objects are identified in an image using one or more neural networks, in which the one or more neural networks are trained using one or more decay parameters.

Using decay parameters for inferencing with neural networks

In various examples, sensor data representative of an image of a field of view of a vehicle sensor may be received and the sensor data may be applied to a machine learning model. The machine learning model may compute a segmentation mask representative of portions of the image corresponding to lane markings of the driving surface of the vehicle. Analysis of the segmentation mask may be performed to determine lane marking types, and lane boundaries may be generated by performing curve fitting on the lane markings corresponding to each of the lane marking types. The data representative of the lane boundaries may then be sent to a component of the vehicle for use in navigating the vehicle through the driving surface.

Real-time detection of lanes and boundaries by autonomous vehicles

In various examples, a deep neural network (DNN) is trained to accurately predict, in deployment, distances to objects and obstacles using image data alone. The DNN may be trained with ground truth data that is generated and encoded using sensor data from any number of depth predicting sensors, such as, without limitation, RADAR sensors, LIDAR sensors, and/or SONAR sensors. Camera adaptation algorithms may be used in various embodiments to adapt the DNN for use with image data generated by cameras with varying parameters—such as varying fields of view. In some examples, a post-processing safety bounds operation may be executed on the predictions of the DNN to ensure that the predictions fall within a safety-permissible range.

Distance to obstacle detection in autonomous machine applications

In various examples, detected object data representative of locations of detected objects in a field of view may be determined. One or more clusters of the detected objects may be generated based at least in part on the locations and features of the cluster may be determined for use as inputs to a machine learning model(s). A confidence score, computed by the machine learning model(s) based at least in part on the inputs, may be received, where the confidence score may be representative of a probability that the cluster corresponds to an object depicted at least partially in the field of view. Further examples provide approaches for determining ground truth data for training object detectors, such as for determining coverage values for ground truth objects using associated shapes, and for determining soft coverage values for ground truth objects.

Object detection and detection confidence suitable for autonomous driving

In various examples, a sequential deep neural network (DNN) may be trained using ground truth data generated by correlating (e.g., by cross-sensor fusion) sensor data with image data representative of a sequences of images. In deployment, the sequential DNN may leverage the sensor correlation to compute various predictions using image data alone. The predictions may include velocities, in world space, of objects in fields of view of an ego-vehicle, current and future locations of the objects in image space, and/or a time-to-collision (TTC) between the objects and the ego-vehicle. These predictions may be used as part of a perception system for understanding and reacting to a current physical environment of the ego-vehicle.

Temporal information prediction in autonomous machine applications

In various examples, a deep neural network (DNN) is trained to accurately predict, in deployment, distances to objects and obstacles using image data alone. The DNN may be trained with ground truth data that is generated and encoded using sensor data from any number of depth predicting sensors, such as, without limitation, RADAR sensors, LIDAR sensors, and/or SONAR sensors. Camera adaptation algorithms may be used in various embodiments to adapt the DNN for use with image data generated by cameras with varying parameters&#x2014;such as varying fields of view. In some examples, a post-processing safety bounds operation may be executed on the predictions of the DNN to ensure that the predictions fall within a safety-permissible range.

In various examples, a deep neural network (DNN) is trained&#x2014;using image data alone&#x2014;to accurately predict distances to objects, obstacles, and/or a detected free-space boundary. The DNN may be trained with ground truth data that is generated using sensor data representative of motion of an ego-vehicle and/or sensor data from any number of depth predicting sensors&#x2014;such as, without limitation, RADAR sensors, LIDAR sensors, and/or SONAR sensors. The DNN may be trained using two or more loss functions each corresponding to a particular portion of the environment that depth is predicted for, such that&#x2014;in deployment&#x2014;more accurate depth estimates for objects, obstacles, and/or the detected free-space boundary are computed by the DNN. In some embodiments, a sampling algorithm may be used to sample depth values corresponding to an input resolution of the DNN from a predicted depth map of the DNN at an output resolution of the DNN.

A communications system has a cellular structure including a base station that is located within a cell of the cellular structure and provides an elevation beamforming transmission based on a set of elevation precoding matrix indicator offsets in an elevation codebook. The communications system also includes user equipment that is located within the cell and coupled to the base station to receive the set of elevation precoding matrix indicator offsets and a set of reference signals to provide channel quality and inter-cell interference measurements, wherein a selected channel quality indicator is based on an increase in channel quality with respect to inter-cell interference at the user equipment and corresponds to one of the set of elevation precoding matrix indicator offsets. A method of operating a communications system having a cellular structure is also provided.

Soft codebook subset restriction for elevation beamforming

A communications system has a cellular structure and the communications system includes a base station that is located within a cell of the cellular structure and employs a Kronecker product of azimuth and elevation precoding vectors for beamforming. Additionally, the communications system includes user equipment that is located within the cell and coupled to the base station to receive a reference channel state information process employing a reference precoding vector for use in a non-reference channel state information process to derive a compensated channel quality indication. A method of operating a communications system is also included.

Channel quality indication compensation for kronecker precoding

Growing community of inventors

Uusimaa, Finland

Pekka Janis